bia-bob

BIA Bob is a Jupyter+LLM-based assistant for interacting with image data and for working on Bio-image Analysis tasks.

Stars: 88

Visit

BIA `bob` is a Jupyter-based assistant for interacting with data using large language models to generate Python code. It can utilize OpenAI's chatGPT, Google's Gemini, Helmholtz' blablador, and Ollama. Users need respective accounts to access these services. Bob can assist in code generation, bug fixing, code documentation, GPU-acceleration, and offers a no-code custom Jupyter Kernel. It provides example notebooks for various tasks like bio-image analysis, model selection, and bug fixing. Installation is recommended via conda/mamba environment. Custom endpoints like blablador and ollama can be used. Google Cloud AI API integration is also supported. The tool is extensible for Python libraries to enhance Bob's functionality.

README:

bia-bob

BIA bob is a Jupyter-based assistant for interacting with data using large language models which generate Python code for Bio-Image Analysis (BIA). It can make use of OpenAI's chatGPT, Google's Gemini, Anthropic's Claude, Github Models Marketplace, Helmholtz' blablador and Ollama. You need an OpenAI API account or a Google Cloud account or a Helmholtz ID account to use it. Using it with Ollama is free but requires running an Ollama server locally.

bob can write short Python code snippets and entire Jupyter notebooks for your image / data analysis workflow.

Disclaimer

bia-bob is a research project aiming at streamlining the design of image analysis workflows. Under the hood it uses artificial intelligence / large language models to generate text and code fulfilling the user's requests. Users are responsible to verify the generated code according to good scientific practice. Some general advice:

If you do not understand what a generated code snippet does, ask %%bob explain this code in detail to a Python beginner: before executing the code.
After Bob generated a data analysis workflow for you, ask %%bob How could I verify this analysis workflow ?. It is good scientific practice to measure the quality of segmentation results for example, or to measure the difference of automated quantitative measurements, in comparison to manual analysis.
If you are not sure if an image analysis workflow is valid, consider asking human experts. E.g. reach out via https://image.sc .

[!CAUTION] When using the OpenAI, Google Gemini, Anthropic, Github Models or any other endpoint via BiA-Bob, you are bound to the terms of service of the respective companies or organizations. The prompts you enter are transferred to their servers and may be processed and stored there. Make sure to not submit any sensitive, confidential or personal data. Also using these services may cost money.

Usage

You can initialize bob like this:

from bia_bob import bob

Code generation

You can ask Bob to generate code like this:

%bob Load blobs.tif and show it

It will then respond with a Python code snippet that you can execute (see full example):

Use %bob if you want to write in the same line and %%bob if you want to write below.

If you want to continue using a variable in the next code cell, you need to specify the name of the variable in the following prompt.

Notebook generation

When asking Bob explicitly to generate a notebook, it will put a new notebook file in the current directory with the generated code (See full example). You can then open it in Jupyter lab.

Notebook modification

You can also ask Bob to modify an existing notebook, e.g. to introduce explanatory markdown cells (See full example):

Furthermore, one can translate Jupyter notebooks to other languages, e.g. by prompting %bob translate the filename.ipynb to <language>.

Writing prompts

You can also ask Bob to write a prompt for you. This can be useful to explore potential strategies for analyzing image data. Note: It might be necessary to modify those prompts, especially when suggested analysis workflows are long and complicated. Shorten suggested prompts to the minimal necessary steps to answer your scientific question. (See full example).

Prompt augmentation

You can add additional information from given Python variables into your prompt using the {variable_name} syntax. With this, the content of the variable will become part of the prompt (full example).

Explaining code

If you are not sure what generated code does, you can ask Bob to explain it to you:

Bug fixing

Bob can fix simple bugs in code you executed. Just add %%fix on top of the cell right after the error happened.

Code documentation

Using the %%doc magic, you can generate documentation for a given code cell.

GPU-acceleration

Using the %%acc magic, you can replace common image processing functions with GPU-accelerated functions. It is recommended to check if the image processing results remain the same. You can see an example in this notebook.

Example notebooks

Examples

Command line interface

You can use bia-bob from the terminal. This is recommended for creating notebooks for example like this:

bia-bob Please create a Jupyter Notebook that opens blobs.tif, segments the bright objects and shows the resulting label image on top of the original image with a curtain.

This can also be used to create other files, e.g. CSV files.

Known issues

If you want to ask bob a question, you need to put a space before the ?.

%bob What do you know about blobs.tif ?

Installation

You can install bia-bob using conda/pip. It is recommended to install it into a conda/mamba environment. If you have never used conda before, please read this guide first.

It is recommended to install bia-bob in a conda-environment together with useful tools for bio-image analysis.

conda env create -f https://github.com/haesleinhuepf/bia-bob/raw/main/environment.yml

You can then activate this environment...

conda activate bob_env

OR install bob into an existing environment:

pip install bia-bob

Setting API keys

For using LLMs from remote service providers, you need to set an API key. API Keys are short cryptic texts such as "proj_sk_asdasdasd" which allow you to log into a remote service without entering your username and password. Many online serives require using API keys for billing; others enable you to use their free services only after obtaining an API key. This also means that you should not share your API key with others. In the following sections, you find links to a couple of LLM services providers that are compatible with bia-bob. After obtaining the key, you need to add it to the enviroment variables of your computer. On Windows, you can do this by 1) searching for "env" in the start menu, 2) clicking on "Edit the system environment variables", 3) clicking on "Environment Variables", 4) clicking on "New" in the "System variables" section and adding a new variable with the name specified below (e.g. OPENAI_API_KEY) and the value of your API key.

On Linux and MacOS, this is typically done by modifying a hidden .bashrc or .zshrc file in the home directory, e.g. like this:

echo "export OPENAI_API_KEY='yourkey'" >> ~/.zshrc

Note: After setting the environment variables, you need to restart your terminal and/or Jupyter Lab to make them work.

See also further instructions on this page.

Using OpenAI / ChatGPT as backend

Create an OpenAI API Key and add it to your environment variables named OPENAI_API_KEY as explained on

You can then initialize Bob like this (optional, as that's default):

from bia_bob import bob
bob.initialize("gpt-4o-2024-08-06", vision_model="gpt-4o-2024-08-06")

Using Anthropic / Claude

Create an Anthropic API Key and add it to your environment variables named ANTHROPIC_API_KEY.

You can then initialize Bob like this:

from bia_bob import bob
bob.initialize(model="claude-3-5-sonnet-20240620", vision_model="claude-3-5-sonnet-20240620")

Using KISSKI / Academic Cloud

You can also apply for an API Key from the German Artificial Intelligence Service Center for Sensible and Critical Infrastructures who operates the ChatAI service.

You can store it in an environment variable named OPENAI_API_KEY and use initialize bob like this:

from bia_bob import bob
bob.initialize(endpoint="https://chat-ai.academiccloud.de/v1", 
               model="meta-llama-3.1-70b-instruct")

Using Github Models Marketplace

If you are using the models from Github Models Marketplace, please create an GITHUB API key (with default settings) and store it for accessing the models in an environment variable named GH_MODELS_API_KEY.

You can then access the models like this:

bob.initialize(
    endpoint='github_models', 
    model='Phi-3.5-mini-instruct')

Using Azure

If you are using the models hosted on Microsoft Azure, please store your API key for accessing the models in an environment variable named AZURE_API_KEY.

You can then access the models like this:

bob.initialize(
    endpoint='azure', 
    model='Phi-3.5-mini-instruct')

Alternatively, you can specify the endpoint directly, too:

bob.initialize(
    endpoint='https://models.inference.ai.azure.com', 
    model='Phi-3.5-mini-instruct')

Using custom endpoints

Custom endpoints can be used as well if they support the OpenAI API. Examples are blablador and ollama. An example is shown in this notebook:

For this, just install the openai backend as explained above (tested version: 1.5.0).

If you want to use ollama and e.g. the codellama model, you must run ollama serve from a separate terminal and then initialize bob like this:

bob.initialize(endpoint='ollama', model='codellama')

If you want to use blablador, which is free for German academics, just get an API key as explained on this page and store it in your environment as BLABLADOR_API_KEY variable.

bob.initialize(
    endpoint='blablador', 
    model='Mistral-7B-Instruct-v0.2')

Custom end points can be used as well, for example like this:

bob.initialize(
    endpoint='http://localhost:11434/v1', 
    model='codellama')

Using Google gemini 1.5 flash / pro

Create a Google API key and store it in the environment variable GOOGLE_API_KEY.

pip install google-generativeai>=0.7.2

You can then initialize Bob like this:

from bia_bob import bob
bob.initialize("gemini-1.5-pro-002")

Using Google's Cloud AI API as backend

Note: This method is deprecated. Use gemini 1.5 as shown above.

pip install google-cloud-aiplatform

(Recommended google-cloud-aiplatform version >= 1.38.1)

To make use of the Google Cloud API, you need to create a Google Cloud account here and a project within the Google cloud (for billing) here. You need to store authentication details locally as explained here. This requires installing Google Cloud CLI. In very short: run the installer and when asked, activate the "Run gcloud init" checkbox. Or run 'gcloud init' from the terminal yourself. Restart the terminal window. After installing Google Cloud CLI, start a terminal and authenticate using:

gcloud auth application-default login

Follow the instructions in the browser. Enter your Project ID (not the name). If it worked the terminal should approximately look like this:

Development

If you want to contribute to bia-bob, you can install it in development mode like this:

git clone https://github.com/haesleinhuepf/bia-bob.git
cd bia-bob
pip install -e .

Extensibility

If you are maintainer of a Python library and want to make BiA-bob aware of functions in your library, you can extend Bob's knowledge using entry-points. Add this to your library setup.cfg:


[options.entry_points]
bia_bob_plugins =
    plugin1 = your_library._bia_bob_plugins:list_bia_bob_plugins

In the above mentioned _bia_bob_plugins.py define this function (and feel free to rename the function and the Python file):

def list_bia_bob_plugins():
    """List of function hints for bia_bob"""
    return """
    * Computes the sum of a and b
    your_library.compute_sum(a:int,b:int) -> int
    
    * Determines the difference between a and b
    your_library.compute_difference(a:int, b:int) -> int
    
    """

Note that the syntax should be pretty much as shown above: A bullet point with a short description and a code-snippet just below. You can also generate the list_bia_bob_plugins function as demonstrated in this notebook. Please only list the most important functions. If the list of all plugins extending BiA-Bob becomes too long, the prompt will exceed the maximum prompt length.

List of known Python libraries that provide extensions to Bob:

(Feel free to extend this list by sending a pull-request)

Similar projects

There are similar projects:

Issues

If you encounter any problems or want to provide feedback or suggestions, please create a thread on image.sc along with a detailed description and tag @haesleinhuepf .

For Tasks:

Click tags to check more tools for each tasks

generate code fix bugs document code accelerate processing explore data

For Jobs:

data scientist software engineer research scientist machine learning engineer bioinformatician

Alternative AI tools for bia-bob

Similar Open Source Tools

bia-bob

github

: 88

obs-cleanstream

CleanStream is an OBS plugin that utilizes AI to clean live audio streams by removing unwanted words and utterances, such as 'uh's and 'um's, and configurable words like profanity. It uses a neural network (OpenAI Whisper) in real-time to predict speech and eliminate unwanted words. The plugin is still experimental and not recommended for live production use, but it is functional for testing purposes. Users can adjust settings and configure the plugin to enhance audio quality during live streams.

github

: 90

obs-cleanstream

CleanStream is an OBS plugin that utilizes real-time local AI to clean live audio streams by removing unwanted words and utterances, such as 'uh' and 'um', and configurable words like profanity. It employs a neural network (OpenAI Whisper) to predict speech in real-time and eliminate undesired words. The plugin runs efficiently using the Whisper.cpp project from ggerganov. CleanStream offers users the ability to adjust settings and add the plugin to any audio-generating source in OBS, providing a seamless experience for content creators looking to enhance the quality of their live audio streams.

github

: 110

torchchat

torchchat is a codebase showcasing the ability to run large language models (LLMs) seamlessly. It allows running LLMs using Python in various environments such as desktop, server, iOS, and Android. The tool supports running models via PyTorch, chatting, generating text, running chat in the browser, and running models on desktop/server without Python. It also provides features like AOT Inductor for faster execution, running in C++ using the runner, and deploying and running on iOS and Android. The tool supports popular hardware and OS including Linux, Mac OS, Android, and iOS, with various data types and execution modes available.

github

: 3.3k

linkedin-api

The Linkedin API for Python allows users to programmatically search profiles, send messages, and find jobs using a regular Linkedin user account. It does not require 'official' API access, just a valid Linkedin account. However, it is important to note that this library is not officially supported by LinkedIn and using it may violate LinkedIn's Terms of Service. Users can authenticate using any Linkedin account credentials and access features like getting profiles, profile contact info, and connections. The library also provides commercial alternatives for extracting data, scraping public profiles, and accessing a full LinkedIn API. It is not endorsed or supported by LinkedIn and is intended for educational purposes and personal use only.

github

: 1.8k

gpt-engineer

GPT-Engineer is a tool that allows you to specify a software in natural language, sit back and watch as an AI writes and executes the code, and ask the AI to implement improvements.

github

: 51.9k

neural

Neural is a Vim and Neovim plugin that integrates various machine learning tools to assist users in writing code, generating text, and explaining code or paragraphs. It supports multiple machine learning models, focuses on privacy, and is compatible with Vim 8.0+ and Neovim 0.8+. Users can easily configure Neural to interact with third-party machine learning tools, such as OpenAI, to enhance code generation and completion. The plugin also provides commands like `:NeuralExplain` to explain code or text and `:NeuralStop` to stop Neural from working. Neural is maintained by the Dense Analysis team and comes with a disclaimer about sending input data to third-party servers for machine learning queries.

github

: 417

langroid-examples

Langroid-examples is a repository containing examples of using the Langroid Multi-Agent Programming framework to build LLM applications. It provides a collection of scripts and instructions for setting up the environment, working with local LLMs, using OpenAI LLMs, and running various examples. The repository also includes optional setup instructions for integrating with Qdrant, Redis, Momento, GitHub, and Google Custom Search API. Users can explore different scenarios and functionalities of Langroid through the provided examples and documentation.

github

: 106

depthai

This repository contains a demo application for DepthAI, a tool that can load different networks, create pipelines, record video, and more. It provides documentation for installation and usage, including running programs through Docker. Users can explore DepthAI features via command line arguments or a clickable QT interface. Supported models include various AI models for tasks like face detection, human pose estimation, and object detection. The tool collects anonymous usage statistics by default, which can be disabled. Users can report issues to the development team for support and troubleshooting.

github

: 927

BentoVLLM

BentoVLLM is an example project demonstrating how to serve and deploy open-source Large Language Models using vLLM, a high-throughput and memory-efficient inference engine. It provides a basis for advanced code customization, such as custom models, inference logic, or vLLM options. The project allows for simple LLM hosting with OpenAI compatible endpoints without the need to write any code. Users can interact with the server using Swagger UI or other methods, and the service can be deployed to BentoCloud for better management and scalability. Additionally, the repository includes integration examples for different LLM models and tools.

github

: 62

mosec

Mosec is a high-performance and flexible model serving framework for building ML model-enabled backend and microservices. It bridges the gap between any machine learning models you just trained and the efficient online service API. * **Highly performant** : web layer and task coordination built with Rust 🦀, which offers blazing speed in addition to efficient CPU utilization powered by async I/O * **Ease of use** : user interface purely in Python 🐍, by which users can serve their models in an ML framework-agnostic manner using the same code as they do for offline testing * **Dynamic batching** : aggregate requests from different users for batched inference and distribute results back * **Pipelined stages** : spawn multiple processes for pipelined stages to handle CPU/GPU/IO mixed workloads * **Cloud friendly** : designed to run in the cloud, with the model warmup, graceful shutdown, and Prometheus monitoring metrics, easily managed by Kubernetes or any container orchestration systems * **Do one thing well** : focus on the online serving part, users can pay attention to the model optimization and business logic

github

: 774

telemetry-airflow

This repository codifies the Airflow cluster that is deployed at workflow.telemetry.mozilla.org (behind SSO) and commonly referred to as "WTMO" or simply "Airflow". Some links relevant to users and developers of WTMO: * The `dags` directory in this repository contains some custom DAG definitions * Many of the DAGs registered with WTMO don't live in this repository, but are instead generated from ETL task definitions in bigquery-etl * The Data SRE team maintains a WTMO Developer Guide (behind SSO)

github

: 176

fal-js

The fal.ai JS client is a robust and user-friendly library for seamless integration of fal serverless functions in Web, Node.js, and React Native applications. Developed in TypeScript, it provides developers with type safety right from the start. The client library is crafted as a lightweight layer atop platform standards like `fetch`, ensuring hassle-free integration into existing codebases and flawless operation across various JavaScript runtimes. The client proxy feature allows secure handling of credentials by using a server proxy for serverless APIs. The repository also includes example Next.js applications for demonstration and integration.

github

: 74

sage

Sage is a tool that allows users to chat with any codebase, providing a chat interface for code understanding and integration. It simplifies the process of learning how a codebase works by offering heavily documented answers sourced directly from the code. Users can set up Sage locally or on the cloud with minimal effort. The tool is designed to be easily customizable, allowing users to swap components of the pipeline and improve the algorithms powering code understanding and generation.

github

: 705

code2prompt

code2prompt is a command-line tool that converts your codebase into a single LLM prompt with a source tree, prompt templating, and token counting. It automates generating LLM prompts from codebases of any size, customizing prompt generation with Handlebars templates, respecting .gitignore, filtering and excluding files using glob patterns, displaying token count, including Git diff output, copying prompt to clipboard, saving prompt to an output file, excluding files and folders, adding line numbers to source code blocks, and more. It helps streamline the process of creating LLM prompts for code analysis, generation, and other tasks.

github

: 965

WindowsAgentArena

Windows Agent Arena (WAA) is a scalable Windows AI agent platform designed for testing and benchmarking multi-modal, desktop AI agents. It provides researchers and developers with a reproducible and realistic Windows OS environment for AI research, enabling testing of agentic AI workflows across various tasks. WAA supports deploying agents at scale using Azure ML cloud infrastructure, allowing parallel running of multiple agents and delivering quick benchmark results for hundreds of tasks in minutes.

github

: 147

For similar tasks

bia-bob

github

: 88

cody

Cody is a free, open-source AI coding assistant that can write and fix code, provide AI-generated autocomplete, and answer your coding questions. Cody fetches relevant code context from across your entire codebase to write better code that uses more of your codebase's APIs, impls, and idioms, with less hallucination.

github

: 2.6k

auto-dev-vscode

AutoDev for VSCode is an AI-powered coding wizard with multilingual support, auto code generation, and a bug-slaying assistant. It offers customizable prompts and features like Auto Dev/Testing/Document/Agent. The tool aims to enhance coding productivity and efficiency by providing intelligent assistance and automation capabilities within the Visual Studio Code environment.

github

: 221

code2prompt

github

: 965

fittencode.nvim

Fitten Code AI Programming Assistant for Neovim provides fast completion using AI, asynchronous I/O, and support for various actions like document code, edit code, explain code, find bugs, generate unit test, implement features, optimize code, refactor code, start chat, and more. It offers features like accepting suggestions with Tab, accepting line with Ctrl + Down, accepting word with Ctrl + Right, undoing accepted text, automatic scrolling, and multiple HTTP/REST backends. It can run as a coc.nvim source or nvim-cmp source.

github

: 83

chatgpt

The ChatGPT R package provides a set of features to assist in R coding. It includes addins like Ask ChatGPT, Comment selected code, Complete selected code, Create unit tests, Create variable name, Document code, Explain selected code, Find issues in the selected code, Optimize selected code, and Refactor selected code. Users can interact with ChatGPT to get code suggestions, explanations, and optimizations. The package helps in improving coding efficiency and quality by providing AI-powered assistance within the RStudio environment.

github

: 310

pandas-ai

PandasAI is a Python library that makes it easy to ask questions to your data in natural language. It helps you to explore, clean, and analyze your data using generative AI.

github

: 12.9k

supersonic

SuperSonic is a next-generation BI platform that integrates Chat BI (powered by LLM) and Headless BI (powered by semantic layer) paradigms. This integration ensures that Chat BI has access to the same curated and governed semantic data models as traditional BI. Furthermore, the implementation of both paradigms benefits from the integration: * Chat BI's Text2SQL gets augmented with context-retrieval from semantic models. * Headless BI's query interface gets extended with natural language API. SuperSonic provides a Chat BI interface that empowers users to query data using natural language and visualize the results with suitable charts. To enable such experience, the only thing necessary is to build logical semantic models (definition of metric/dimension/tag, along with their meaning and relationships) through a Headless BI interface. Meanwhile, SuperSonic is designed to be extensible and composable, allowing custom implementations to be added and configured with Java SPI. The integration of Chat BI and Headless BI has the potential to enhance the Text2SQL generation in two dimensions: 1. Incorporate data semantics (such as business terms, column values, etc.) into the prompt, enabling LLM to better understand the semantics and reduce hallucination. 2. Offload the generation of advanced SQL syntax (such as join, formula, etc.) from LLM to the semantic layer to reduce complexity. With these ideas in mind, we develop SuperSonic as a practical reference implementation and use it to power our real-world products. Additionally, to facilitate further development we decide to open source SuperSonic as an extensible framework.

github

: 2.2k

For similar jobs

NoLabs

NoLabs is an open-source biolab that provides easy access to state-of-the-art models for bio research. It supports various tasks, including drug discovery, protein analysis, and small molecule design. NoLabs aims to accelerate bio research by making inference models accessible to everyone.

github

: 75

OpenCRISPR

OpenCRISPR is a set of free and open gene editing systems designed by Profluent Bio. The OpenCRISPR-1 protein maintains the prototypical architecture of a Type II Cas9 nuclease but is hundreds of mutations away from SpCas9 or any other known natural CRISPR-associated protein. You can view OpenCRISPR-1 as a drop-in replacement for many protocols that need a cas9-like protein with an NGG PAM and you can even use it with canonical SpCas9 gRNAs. OpenCRISPR-1 can be fused in a deactivated or nickase format for next generation gene editing techniques like base, prime, or epigenome editing.

github

: 253

ersilia

The Ersilia Model Hub is a unified platform of pre-trained AI/ML models dedicated to infectious and neglected disease research. It offers an open-source, low-code solution that provides seamless access to AI/ML models for drug discovery. Models housed in the hub come from two sources: published models from literature (with due third-party acknowledgment) and custom models developed by the Ersilia team or contributors.

github

: 211

ontogpt

OntoGPT is a Python package for extracting structured information from text using large language models, instruction prompts, and ontology-based grounding. It provides a command line interface and a minimal web app for easy usage. The tool has been evaluated on test data and is used in related projects like TALISMAN for gene set analysis. OntoGPT enables users to extract information from text by specifying relevant terms and provides the extracted objects as output.

github

: 584

bia-bob

github

: 88

Scientific-LLM-Survey

Scientific Large Language Models (Sci-LLMs) is a repository that collects papers on scientific large language models, focusing on biology and chemistry domains. It includes textual, molecular, protein, and genomic languages, as well as multimodal language. The repository covers various large language models for tasks such as molecule property prediction, interaction prediction, protein sequence representation, protein sequence generation/design, DNA-protein interaction prediction, and RNA prediction. It also provides datasets and benchmarks for evaluating these models. The repository aims to facilitate research and development in the field of scientific language modeling.

github

: 161

polaris

Polaris establishes a novel, industry‑certified standard to foster the development of impactful methods in AI-based drug discovery. This library is a Python client to interact with the Polaris Hub. It allows you to download Polaris datasets and benchmarks, evaluate a custom method against a Polaris benchmark, and create and upload new datasets and benchmarks.

github

: 85

awesome-AI4MolConformation-MD

The 'awesome-AI4MolConformation-MD' repository focuses on protein conformations and molecular dynamics using generative artificial intelligence and deep learning. It provides resources, reviews, datasets, packages, and tools related to AI-driven molecular dynamics simulations. The repository covers a wide range of topics such as neural networks potentials, force fields, AI engines/frameworks, trajectory analysis, visualization tools, and various AI-based models for protein conformational sampling. It serves as a comprehensive guide for researchers and practitioners interested in leveraging AI for studying molecular structures and dynamics.

github

: 107